MIT researchers are designing a machine to replicate a dog’s sense of smell

We recently shared a story about how robots can alleviate loneliness much like a pet. It turns out that robotics can replicate other great things canines do. Researchers from the Massachusetts Institute of Technology (MIT) are developing a machine that can emulate a dog’s acute sense of smell. 

After years of development, the researchers successfully trained a robot to replicate a dog’s sense of smell by using a machine learning algorithm. The machine was able to successfully identify tiny traces of molecules with sensitivity 200 times more than that of a dog’s nose but fell short when it came to identifying what the scents were. 

Now, the scientists are working with a more precise algorithm and 50 urine samples, taken from prostate cancer patients as well as cancer-free patients, to teach the machine to differentiate between sick and healthy individuals. So far, the machine has achieved a prostate cancer recognition accuracy of 70 percent. Moving forwards, the team hopes to perfect the technology and make it accessible in a small, handheld format for use in medical settings. If successful, the machine could offer the health and safety resources of highly trained dogs in a simple device. 

Dogs have been used for years to sniff out contraband and potential weapons. More recently, their skills have been deployed to protect wildlife and even slow the spread of Covid-19. There will likely always be a place for these service dogs, but making their skills available in a machine would make detection technology far more widely available and even expand it to new potential applications.

Solution News Source

MIT researchers are designing a machine to replicate a dog’s sense of smell

We recently shared a story about how robots can alleviate loneliness much like a pet. It turns out that robotics can replicate other great things canines do. Researchers from the Massachusetts Institute of Technology (MIT) are developing a machine that can emulate a dog’s acute sense of smell. 

After years of development, the researchers successfully trained a robot to replicate a dog’s sense of smell by using a machine learning algorithm. The machine was able to successfully identify tiny traces of molecules with sensitivity 200 times more than that of a dog’s nose but fell short when it came to identifying what the scents were. 

Now, the scientists are working with a more precise algorithm and 50 urine samples, taken from prostate cancer patients as well as cancer-free patients, to teach the machine to differentiate between sick and healthy individuals. So far, the machine has achieved a prostate cancer recognition accuracy of 70 percent. Moving forwards, the team hopes to perfect the technology and make it accessible in a small, handheld format for use in medical settings. If successful, the machine could offer the health and safety resources of highly trained dogs in a simple device. 

Dogs have been used for years to sniff out contraband and potential weapons. More recently, their skills have been deployed to protect wildlife and even slow the spread of Covid-19. There will likely always be a place for these service dogs, but making their skills available in a machine would make detection technology far more widely available and even expand it to new potential applications.

Solution News Source

SIGN UP

TO GET A Free DAILY DOSE OF OPTIMISM


We respect your privacy and take protecting it seriously. Privacy Policy